This invention relates to ignition devices for engines of the kind in which a fuel-air mixture is introduced into a combustion chamber where it is compressed, ignited, expanded and exhausted in a repetitive cycle. It is applicable to Otto cycle engines, including two and four cycle piston engines, free piston engines, and rotary engines, such as Wankel engines. The invention is particularly concerned with ignition cell devices of the kind which are capable of performing the ignition function in a running engine without the use of an externally-supplied, timed electrical spark, which has been the type of ignition system most widely used heretofore.
Despite their nearly universal employment, electrical ignition systems for internal combustion engines have well-recognized disadvantages, One of these is the cost of the ignition equipment itself. Another is the necessity for maintenance of the parts of the electrical system. Such a system normally includes a number of small moving parts that wear more rapidly than the more massive moving parts of the engine itself, and electrical components which deteriorate through exposure to heat, moisture, engine oil and oil residue.
Electrical spark ignition systems broadcast in commonly used radio frequencies, and special steps must be taken to shield them or otherwise prevent this radio interference.
Another disadvantage is that the added parts involved in an electrical ignition system decrease the reliability and longevity of the engine operation. For critical applications, such as in aircraft engines, it has become mandatory to use a dual electrical system with complete duplication of parts and equipment, in order to gain some assurance of reliability.
One common source of electrical ignition system unreliability is the fouling of spark plugs caused by lubricating oil being forced past seals (e.g. piston rings) into the combustion chamber and incompletely burning there to leave a deposit on the spark plug electrodes. Organometallic deposits derived from fuel additives are another source of spark plug fouling.
The problem of electrical ignition system reliability is much more severe in an engine equipped with a catalytic exhaust gas converter, as is presently contemplated for general adoption to reduce air pollution, than it is in an engine not so equipped. Failure of a single spark plug in a multi-cylinder engine will ordinarily not cause an engine stoppage, but it will result in the pumping of a large quantity of unburned fuel (mixed with air) through the cylinder having the failed plug into the catalyst chamber. There the mixture will burn, and this burning of a larger-than-planned-for quantity of fuel in a converter designed to handle relatively minute amounts of unburned fuel per unit time will quickly cause the catalyst to overheat, break down, form a powder, and blow out the exhaust pipe. In less than five minutes an entire expensive catalyst charge can be destroyed in this manner.
Other disadvantages are inherent in electrical ignition systems. The timing of such systems, even when performed through solid state electronic circuits, is essentially mechanical, that is, the time of initiation of a spark at a spark plug is determined by the position of the mechanical parts of the engine and the combustion chamber conditions which ought to exist with the mechanical parts in that position, rather than being determined by the precombustion conditions actually existing in the engine. Wear or maladjustment in the mechanical timing system causes mistiming of the spark. In addition, experience has shown that spark timing should be varied with engine speed, and this, too, is accomplished mechanically, with a similar liability to mistiming through wear or maladjustment.
Another disadvantage is that a spark plug, no matter how it is designed, results in initiation of the burning of the fuel in the combustion chamber at a very localized point, a fact having numerous undesirable implications, including incomplete combustion, the need for intense care in the design of the combustion chamber space, and cooling problems. One undesirable effect attributable to the very localized commencement of ignition inherent in a spark system is cycle-to-cycle variation in the performance of a single cylinder, reflected in its p-v diagram.
Ignition cells or cavities associated with combustion chambers in internal combustion engines have been proposed in the past. See, for example, U.S. Pat. No. 2,996,056 to Vierling, U.S. Pat. No. 2,279,709 to Kite and U.S. Pat. No. 3,481,317 to Hughes and DePalma. However, such devices have not come into widespread use because of existing limitations. While it has been possible to provide an ignition cell or cavity which functions well in a given engine at constant speed and load, it has not thus far been possible to make cells which perform well over a wide range of engine and load conditions. Nor has it been possible to generalize the design parameters or criteria of such cells so that in the present state of the art the provision of a cell for a given engine is almost entirely a matter of "cut and try".
Another limitation of existing ignition cells is that they will not start an engine, inamuch as they do not become functional until heated up by heat from the combustion chamber of the engine. The before-mentioned patent to Vierling proposed to overcome this limitation by resistive electrical heating (externally supplied) of the walls of ignition cells. Absent this expedient, in the present state of the art, ignition cells do not permit elimination of the conventional electrical ignition system of an engine, since it is needed at least for starting of the engine and warm-up.